Container (SDH)
How Containers Map PDH Tributaries Into SDH
The container is the lowest layer of the SDH adaptation stack defined in ITU-T G.707. An incoming asynchronous tributary, for example a 2.048 Mbit/s E1, cannot be dropped directly into a synchronous frame because its clock is independent of the SDH network clock and may drift within plus or minus 50 ppm. The container solves this by defining a structure that runs slightly faster than the nominal tributary, padding the difference with fixed stuff bits and a small number of justification bits that can be flipped on a per-multiframe basis. A C-12 runs at 2.176 Mbit/s to carry a 2.048 Mbit/s E1, leaving 128 kbit/s of headroom for stuffing and justification control.
Adding path overhead (POH) to a container produces a virtual container. The POH bytes carry a BIP-8 parity check (B3 for higher-order, V5 bit-2 for lower-order), a trail trace identifier, and a signal label that declares the mapping type. The VC, not the bare container, is the unit that survives end to end across the network; intermediate nodes never disassemble it, they only adjust pointers. Lower-order VCs are bundled through tributary unit groups (TUG-2, then TUG-3) and finally into a single VC-4 that fills the entire payload of an STM-1, so 63 VC-12 containers or 3 VC-3 containers ride inside one 155.52 Mbit/s frame.
For RF transport this matters because a microwave radio terminating an SDH section must reproduce every container bit-exactly. Any mapping jitter introduced when the radio recovers and re-justifies the tributary has to fall within the G.823/G.825 wander masks, otherwise downstream equipment will log jitter errored seconds even when the radio path itself is error free.
Container Rates and Justification
RC = (bytes per frame) × 8 bits × 8000 frames/s
C-12 example (lower-order):
RC-12 = 34 bytes × 8 × 8000 ≈ 2.176 Mbit/s (carries 2.048 Mbit/s E1)
C-4 example (higher-order):
RC-4 = 2340 bytes × 8 × 8000 ≈ 149.76 Mbit/s (carries 139.264 Mbit/s E4)
Gross headroom (mostly fixed stuffing):
ΔR = RC − Rtrib = 2.176 − 2.048 = 0.128 Mbit/s (≈ 6.25% of a C-12, almost all fixed stuff and overhead)
Variable justification capacity (sets the clock-tracking range):
offset range = ±(Rjust / Rtrib), where Rjust is the rate of the variable justification opportunities ≈ ±50 ppm (per ITU-T G.823)
Where RC = container rate, Rtrib = nominal tributary rate, Rjust = capacity of the variable justification bits only. The large ΔR is overwhelmingly fixed stuffing; only the few variable justification bits per multiframe absorb clock offset, which is why the tracking range is ±50 ppm rather than the full 6.25% of ΔR. The 8000 frames/s comes from the 125 μs SDH frame period. STM-1 = 9 × 270 bytes × 8 × 8000 = 155.52 Mbit/s.
SDH Container Types
| Container | Tributary Carried | Tributary Rate | Container Rate | Virtual Container | Per STM-1 |
|---|---|---|---|---|---|
| C-11 | DS1 (T1) | 1.544 Mbit/s | 1.600 Mbit/s | VC-11 | 84 |
| C-12 | E1 | 2.048 Mbit/s | 2.176 Mbit/s | VC-12 | 63 |
| C-2 | DS2 | 6.312 Mbit/s | 6.784 Mbit/s | VC-2 | 21 |
| C-3 | E3 / DS3 | 34.368 / 44.736 Mbit/s | 48.384 Mbit/s | VC-3 | 3 |
| C-4 | E4 | 139.264 Mbit/s | 149.76 Mbit/s | VC-4 | 1 |
Frequently Asked Questions
What is the difference between a container and a virtual container in SDH?
A container (C-n) is the raw fixed-rate structure holding the mapped tributary, for example a C-12 carries a justified 2.048 Mbit/s E1 inside a 2.176 Mbit/s structure. A virtual container (VC-n) adds path overhead (B3 BIP-8, J1/J2 trace, signal label) so it can be monitored end to end. The bare container never travels alone; it always rides inside its VC, which a pointer positions within the higher-order structure.
Which SDH container carries an E1 versus a DS3?
A 2.048 Mbit/s E1 maps into a C-12 (becoming a VC-12), while both the 34.368 Mbit/s E3 and the 44.736 Mbit/s DS3 map into a C-3 (becoming a VC-3). The 139.264 Mbit/s E4 maps into a C-4. Lower-order containers are grouped through TUG-2 and TUG-3 into a VC-4, so one STM-1 can carry 63 VC-12 or 3 VC-3 containers.
Why does a container run faster than the tributary it carries?
The extra rate absorbs frequency offset between the asynchronous PDH source and the SDH network clock. A C-12 at 2.176 Mbit/s leaves room above the 2.048 Mbit/s E1 for fixed stuff bits plus justification opportunities, letting the mapping track a source drifting within ±50 ppm without data loss. The cost is mapping jitter that the desynchronizer phase-locked loop must filter at the receiving end.